• 한국항행학회논문지
• /
• 제25권4호
• /
• pp.299-304
• /
• 2021

기존의 사람 행동 인식 시스템은 웨어러블 센서, 카메라와 같은 장치를 통해 행동을 탐지하였다. 그러나, 이와 같은 방법들은 추가적인 장치와 비용이 요구되고, 특히 카메라 장치의 경우 사생활 침해 문제가 발생한다. 이미 설치되어 있는 WiFi 신호를 사용한다면 해당 문제를 해결할 수 있다는 장점이 있다. 본 논문에서는 WiFi 신호의 채널 상태 정보를 활용한 CNN 기반 사람 행동 인식 시스템을 제안하고, 가속 하드웨어 구조 설계 및 구현 결과를 제시한다. 해당 시스템은 실내 공간에서 학습 중 나타날 수 있는 네 가지 행동에 대해 정의하였고, 그에 대한 WiFi의 채널 상태 정보를 CNN으로 분류하여 평균 정확도는 91.86%를 보였다. 또한, 가속화를 위해 CNN 분류기에서 연산량이 가장 많은 완전 연결 계층에 대한 가속 하드웨어 구조 설계 결과를 제시하였다. FPGA 디바이스 상에서 성능 평가 결과, 기존 software 기반 시스템 대비 4.28배 빠른 연산 시간을 보임을 확인하였다.

• Journal of Positioning, Navigation, and Timing
• /
• 제11권3호
• /
• pp.217-227
• /
• 2022

In order to provide a location-based services regardless of indoor or outdoor space, it is important to provide position information of the terminal regardless of location. Among the wireless/mobile communication resources used for this purpose, Long Term Evolution (LTE) signal is a representative infrastructure that can overcome spatial limitations, but the positioning method based on the location of the base station has a disadvantage in that the accuracy is low. Therefore, a fingerprinting technique, which is a pattern recognition technology, has been widely used. The simplest yet widely applied algorithm among Fingerprint positioning technologies is k-Nearest Neighbors (kNN). However, in the kNN algorithm, it is difficult to find the optimal K value with the lowest positioning error for each location to be estimated, so it is generally fixed to an appropriate K value and used. Since the optimal K value cannot be applied to each estimated location, therefore, there is a problem in that the accuracy of the overall estimated location information is lowered. Considering this problem, this paper proposes a technique for adaptively varying the K value by using a Convolutional Neural Network (CNN) model among Artificial Neural Network (ANN) techniques. First, by using the signal information of the measured values obtained in the service area, an image is created according to the Physical Cell Identity (PCI) and Band combination, and an answer label for supervised learning is created. Then, the structure of the CNN is modeled to classify K values through the image information of the measurements. The performance of the proposed technique is verified based on actual data measured in the testbed. As a result, it can be seen that the proposed technique improves the positioning performance compared to using a fixed K value.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제5권7호
• /
• pp.159-172
• /
• 2016

본 연구는 복층 건물에서 실내 재난상황이 발생했을 때 적절한 대피경로들을 산출하기 위한 새로운 알고리즘을 제안한다. 제안된 방안은 Disaster Evacuation Graph(DEG)를 도입하여, 탐색시간을 크게 단축시킨다. 또한, 대피 경로 상의 인원수용능력과 그에 따른 인원 분산을 동시에 고려한다. 경로를 탐색하는 과정은 크게 두 단계로 구성된다. 이 때 각 단계는 Horitzontal Tiering(HT)과 Vertical Tiering(VT) 단계라 한다. HT 단계에서는 특정 층 임의의 공간으로부터 계단으로 향하는 가능한 최적의 경로를 산출한다. 그리고 모든 층에 대해 HT 단계를 수행한 이후, 각 층의 모든 공간으로부터 건물 입구나 옥상과 같은 안전지대로 향하는 경로를 결정하기 위해, VT 단계에서 HT 단계의 결과들을 통합한다. 이러한 단계별 과정에서 각각, 층 내 모든 공간으로부터 계단으로 향하는 경로와 각 계단으로부터 안전지대로 향하는 경로가 산출된다. 그리고 이러한 과정에서 경로를 탐색하기 위한 그래프의 범위를 결정하기 위해 티어링(tiering) 기법이 이용된다. 제안된 알고리즘의 성능을 평가하기 위하여, 빠른 탐색과 인원 분산을 위해 기존에 연구되었던 경로탐색 알고리즘들과, 경로탐색에 소요된 시간 및 사용자 대피시간을 비교한다. 제안된 알고리즘은 기존에 비해 더 짧은 시간 내에 빠르게 대피할 수 있는 경로를 산출한다. 특히 경로탐색을 위한 시간복잡도 성능이 뛰어나기 때문에, 높은 층으로 구성된 큰 규모의 건물에서 유용하게 활용할 수 있다.

• 제어로봇시스템학회논문지
• /
• 제13권11호
• /
• pp.1092-1098
• /
• 2007

A mobile robot localization problem can be classified into following three sub-problems as an observation likelihood model, a motion model and a filtering technique. So far, we have developed the range sensor based, integrated localization scheme, which can be used in human-coexisting real environment such as a science museum and office buildings. From those experiences, we found out that there are several significant issues to be solved. In this paper, we focus on three key issues, and then illustrate our solutions to the presented problems. Three issues are listed as follows: (1) Investigation of design requirements of a desirable observation likelihood model, and performance analysis of our design (2) Performance evaluation of the localization result by computing the matching error (3) The semi-global localization scheme to deal with localization failure due to abrupt wheel slippage In this paper, we show the significance of each concept, developed solutions and the experimental results. Experiments were carried out in a typical modern building environment, and the results clearly show that the proposed solutions are useful to develop practical and integrated localization schemes.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2008년도 학술대회 1부
• /
• pp.510-516
• /
• 2008

In ubiquitous environment, there are no limits to utilize networks whenever and wherever you are. That pervasive networks are combined with the information devices and then create various services such as car navigation of LBS(location based service) and DMB(Digital Multimedia Broadcasting). As these kinds of services are getting more various, the complexity is also getting higher and ultimately the convergence will make people feel frustrated. One of the solutions is Context-Awareness[1] technology. User interface with context-awareness filters unnecessary information and prevents users from being blocked due to the massive information. In this paper, we describe the seamless service system based on location-awareness, which is a type of context-awareness. We developed the system based on UPnP AV Framework. This system provides the automatic terminal device selection for the nomadic user. The system establishes new connections for the ongoing streaming playback session with the new AV devices without substantial loss of playback so that the user can enjoy the seamless service. The AV device selection based on the user's location needs no user's intervention or notification so it achieves the improvement of usability and complexity.

• International Journal of Aeronautical and Space Sciences
• /
• 제11권2호
• /
• pp.69-79
• /
• 2010

This paper describes the experimental framework for the control system design and validation of a rotorcraft unmanned aerial vehicle (UAV). Our approach follows the general procedure of nonlinear modeling, linear controller design, nonlinear simulation and flight test but uses an indoor-installed multi-camera system, which can provide full 6-degree of freedom (DOF) navigation information with high accuracy, to overcome the limitation of an outdoor flight experiment. In addition, a 3-DOF flying mill is used for the performance validation of the attitude control, which considers the characteristics of the multi-rotor type rotorcraft UAV. Our framework is applied to the design and mathematical modeling of the control system for a quad-rotor UAV, which was selected as the test-bed vehicle, and the controller design using the classical proportional-integral-derivative control method is explained. The experimental results showed that the proposed approach can be viewed as a successful tool in developing the controller of new rotorcraft UAVs with reduced cost and time.

• 제어로봇시스템학회논문지
• /
• 제16권12호
• /
• pp.1226-1232
• /
• 2010

A door can serve as a feature for place classification and localization for navigation of a mobile robot in indoor environments. This paper proposes a door detection method based on the recognition of various door handles using the general Hough transform (GHT) and support vector machine (SVM). The contour and color histogram of a door handle extracted from the database are used in GHT and SVM, respectively. The door recognition scheme consists of four steps. The first step determines the region of interest (ROI) images defined by the color information and the environment around the door handle for stable recognition. In the second step, the door handle is recognized using the GHT method from the ROI image and the image patches are extracted from the position of the recognized door handle. In the third step, the extracted patch is classified whether it is the image patch of a door handle or not using the SVM classifier. The door position is probabilistically determined by the recognized door handle. Experimental results show that the proposed method can recognize various door handles and detect doors in a robust manner.

• Journal of Electrical Engineering and Technology
• /
• 제8권3호
• /
• pp.658-665
• /
• 2013

While the chirp signal is extensively used in radar and sonar systems for target decision in wireless communication systems, it has not been widely used for positioning in indoor environments. Recently, the IEEE 802.15.4a standard has adopted the chirp spread spectrum (CSS) as an underlying technique for low-power and low-complexity precise localization. Chirp signal based ranging solutions have been established and deployed but their ranging performance has not been analyzed in multipath environments. This paper presents a ranging performance analysis of a chirp signal and suggests a method to suppress multipath error by using a type of non-linear chirp signal. Multipath ranging performance is evaluated using a conventional linear chirp signal and the proposed non-linear chirp signal. We verify the feasibility of both methods using two-ray multipath model simulation. Our results demonstrate that the proposed non-linear chirp signal can successfully suppress the multipath error.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.923-924
• /
• 2006

In this paper, we propose the sensor fusing method for the obstacle avoidance of guiding robot for the visually impaired In our system, we acquire obstacles distances information using ultrasonic sensors, and its width is acquired by image sensor. We also compute avoidance angle using are distance and width information gained by sensor. After the robot avoid the obstacle by computed angle, the robot returns to its original path using odometry. The robot consists of the SA1110-based controller, sensory part using sonar array and image sensor, and motion part using differential drive for climbing stairs. This system use the embedded linux for OS, and also is developed by the QT/Embedded for GUI.

• 전기학회논문지
• /
• 제58권2호
• /
• pp.415-421
• /
• 2009

TOA(time-of-arrival) and TDOA(time-difference-of-arrival) positioning techniques are commonly used in many radio-navigation systems. From the literature, it is known that the position estimate and error covariance matrix of TDOA obtained by GN(Gauss-Newton) method is exactly the same as that of TOA when the error source of the range measurement is only an IID white Gaussian noise. In case of geo-location and indoor positioning, however, multi-path or NLOS(non-line-of-sight) error is frequently appeared in range measurements. Though its occurrence is random, the multipath acts like a bias for a stationary user if it occurs. This paper presents the comparisons of error characteristics between TOA and TDOA positioning in presence of multi-path or NLOS error. It is analytically shown that the position estimate of TDOA is exactly the same as that of TOA even when bias errors are included in range measurements with different magnitudes. By computer simulation, position estimation error and error distribution are analyzed in presence of range bias errors.